The role of pain modulation in non-suicidal self-injury

Abstract

Individuals with non-suicidal self-injury (NSSI) behavior tend to report feeling little or no pain when they self-injure. Moreover, in laboratory studies the NSSI population tends to demonstrate reduced sensitivity to painful stimuli. There is reason to believe that hypoalgesia could be a risk factor for developing and maintaining NSSI. Many theories have been proposed to explain the reduced sensitivity to pain in the NSSI population; some examples are dissociation, self-critical cognitive style, and low levels of endogenous opioids. However, the evidence supporting these theories are sparse. To understand why the NSSI population experiences less pain, there is a need for a better understanding of how individuals with NSSI process pain. We wanted to use methods that have been developed to study pain modulation in individuals with long-term pain to characterize the pain modulation system of women with ongoing NSSI. Our general hypothesis was that women with NSSI have a hyper-effective pain modulation system that inhibits pain to a greater extent and facilitates pain to a lesser extent, compared to women without NSSI.

In Study I, a non-clinical population (N = 62) was recruited to test a pain testing protocol in order to produce offset analgesia (OA) and onset hyperalgesia (OH). Small deviations in a painful thermal stimulation have been found to produce disproportional hypoalgesic (OA) and hyperalgesic (OH) responses. Different stimulus ranges (±1°C and ±2°C) were included in the protocol to study the dynamic relation between heat and pain. The study was composed of two identical experiments. In experiment 1, we produced OA and OH responses, using ±2°C but not ±1°C. In experiment 2, we only produced OA responses, but no OH responses.

Study II investigated if it was possible to induce sensory attenuation of pain in a non-clinical population (N = 40) by comparing self-administered pressure pain threshold to experimenter-administrated pressure pain threshold, using an algometer. An experimental condition, where the participants imagined that they pressed the algometer, was also included in the study, to examine if sensory attenuation could be induced with the help of imagery. Self-administered pressure was found to be less painful, compared to experimenter-administered pressure. Moreover, imagined self-administered pressure was also experienced as less painful than experimenter-administered pressure. Self-induced sensory attenuation of pain could be a factor in explaining hypoalgesia during NSSI.

Study III consisted of an extensive battery of pain tests in order to study pain modulation in a sample of women with NSSI (N = 41) and an age-matched control group, consisting of healthy women (N = 40). The study also included a simple pain test combined with fMRI. We found that the NSSI group demonstrated higher pressure and heat pain thresholds, compared to the control group. The NSSI group also demonstrated a larger conditioned pain modulation (CPM) effect, compared to the control group. CPM is a test based on the principle pain inhibits pain, and is a measure of central down-regulation of pain. We found no difference between the groups regarding temporal summation of pain, a measure of pain facilitation, or in heat pain tolerance. Tonic painful heat stimulation produced a larger hemodynamic response in primary and secondary somatosensory cortex in the NSSI group, compared to the control group.

In Study IV, we used the combined OA/OH protocol that was evaluated in Study I to study pain modulation in women with NSSI (N = 37) and controls (N = 39). The OA/OH protocol was combined with fMRI. Across groups, both the OA and the OH responses were significant. We also found a difference between the groups regarding the OH response, as the NSSI group demonstrated a weaker OH response, compared to the control group. The OH response was associated with a hemodynamic response in the primary somatosensory cortex, across groups, which suggests that the nociceptive signal was upregulated before reaching the brain.

In line with our main hypothesis, we found that the NSSI group inhibited pain to a greater extent (CPM in Study III) and facilitated pain to a lesser extent (OH in Study IV), compared to the control group. These results suggest that women with NSSI have a hyper-effective pain modulation system. There were also results that did not support our main hypothesis; the NSSI group did not demonstrate weaker pain facilitation when tested with the temporal summation protocol (Study III) or stronger inhibition associated with OA (Study IV). An explanation could be that different pain tests measure different aspects of pain modulation and only certain pain modulation mechanisms are affected in the NSSI population. The studies of this thesis provide evidence that the previous findings of hypoalgesia in the NSSI population does not reflect response bias but is rooted in how the nervous system modulates nociceptive signals.